Skip to content

refactor(insights): port Analyst to Nooa #1097

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
BrianNewsom merged 4 commits into from
Aug 6, 2026
Merged

refactor(insights): port Analyst to Nooa #1097

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d539a46
refactor(insights): port analyst to Nooa
BrianNewsom Aug 5, 2026
d65cea1
fix(insights): address Nooa migration review feedback
BrianNewsom Aug 5, 2026
534340f
fix(insights): require secure observability endpoints
BrianNewsom Aug 5, 2026
da10a35
chore: merge main and refresh lockfile
BrianNewsom Aug 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 4 additions & 5 deletions plugins/nemo-insights/examples/research-agent/tests/test_analyst_e2e.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,10 +20,9 @@
- ``NVIDIA_API_KEY`` and ``TAVILY_API_KEY`` — for the research agent's NIM
model + Tavily search. Read from the example's ``.env`` (or the shell).
- ``INFERENCE_API_KEY`` — the ``sk-...`` NVIDIA Inference Gateway virtual key
for the analyst's Claude Opus (served over the Anthropic wire format). The
analyst's LLM ``base_url`` is pinned in
:mod:`nemo_insights_plugin.analyst.agent`, so no base-url override is
required.
for the analyst's Claude Opus. Nooa routes it through LiteLLM using the
gateway's OpenAI-compatible API. The model and base URL are pinned in
:mod:`nemo_insights_plugin.analyst.model_config`, so no override is required.
- Docker — required to auto-start ClickHouse if one isn't already at
``NMP_INTAKE_CLICKHOUSE_URL`` (default ``http://localhost:8123``). A missing
Docker daemon fails the test.
Expand Down Expand Up @@ -464,7 +463,7 @@ def test_analyst_creates_insight_end_to_end(platform_server: str) -> None: # no
f"--- analyst stdout ---\n{result.stdout}\n--- analyst stderr ---\n{result.stderr}"
)

# 6. The analyst's own Pydantic AI spans should be queryable in Intake.
# 6. The analyst's own Nooa spans should be queryable in Intake.
analyst_spans = _wait_for_analyst_spans(SPAN_VISIBLE_TIMEOUT_S)
assert analyst_spans, (
f"expected Intake spans for {ANALYST_OBSERVABILITY_AGENT_NAME}; "
Expand Down
6 changes: 2 additions & 4 deletions plugins/nemo-insights/pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,18 +2,16 @@
name = "nemo-insights-plugin"
version = "0.1.0"
description = "Agent telemetry analysis and persistent insights for NeMo Platform."
requires-python = ">=3.11,<3.15"
requires-python = ">=3.12,<3.14"
dependencies = [
"fastapi>=0.115",
"genai-prices==0.0.62",
"httpx",
"nemo-platform",
"nemo-platform-plugin",
"nooa",
"opentelemetry-exporter-otlp>=1.42.1",
"opentelemetry-sdk>=1.42.1",
"pydantic>=2.10.6",
"pydantic-ai-harness[code-mode]==0.3.0",
"pydantic-ai-slim[anthropic]==1.105.0",
"pyyaml>=6.0.3",
"typer>=0.20.0",
"tzdata==2026.2",
Expand Down
Loading
Loading